Dan Shiebler

Deep neural networks are some of the most powerful learning algorithms that have ever been developed. Unfortunately, they are also some of the most complex. The hierarchical non-linear transformations that neural networks apply to data can be nearly impossible to understand. This problem is exacerbated by the non-determinism of neural... [Read More]

Let’s talk about making assumptions. Each time you form an opinion on something, you don’t re-examine each of the beliefs that contributed to that opinion. For example, when you order food at a restaurant, you pick the foods that you think will taste good. You don’t spend hours questioning the... [Read More]

One of my favorite classes in college was Abstract Algebra. It was my first non-CS non-elementary math class, and I took it because I knew I didn’t want to graduate without having broadened my mathematical horizons beyond CS. As it turned out, Abstract Algebra felt pretty familiar. Many of the... [Read More]

Welcome to part 2 of the Computation Theory series! In the last post, we discussed decidability and the halting problem. In this post we will talk about complexity classes. Time Complexity If you’re not used to reasoning about algorithms and their complexity, there is an enormous amount of great material... [Read More]

When you think about the most important fields of mathematics for a data scientist to know, you’ll almost certainly name calculus, probability/statistics, and linear algebra. These fields form the underpinnings of the most popular data analysis, signal processing and machine learning techniques. But in my experience, I’ve found that there... [Read More]